Open-source platform for 3D AI agents. Turn text or a photo into a rigged, animated GLB avatar, give it an LLM brain, memory and a wallet, and embed it anywhere with one web component. Remote MCP server, x402 payments, Three.js, browser-native.
JavaScript 3d-generationai-agentsanimationavatardigital-humangenerative-aiglbgltfimage-to-3dllmmcpmcp-servermodel-context-protocolsolanatext-to-3dthreejsweb-componentswebglwebxrx402
docs(sniper): stop naming a decommissioned model as the free-chain example
nirholasb0144d240c30
diff --git a/api/_lib/sniper-loops-health.js b/api/_lib/sniper-loops-health.jsindex f211196ce..41c280366 100644--- a/api/_lib/sniper-loops-health.js+++ b/api/_lib/sniper-loops-health.js@@ -143,7 +143,8 @@ export function findWalletlessArms(rows) {* `fallback:` prefix never reaches that column. What the worker's judge writes* on the two paths (workers/agent-sniper/llm-judge.js):* OpenRouter answered -> answered_by === model (same string, same variable)- * free-chain fallback -> answered_by = the raw free model name ("llama-3.1-8b-instant")+ * free-chain fallback -> answered_by = the raw free model name (whichever rung in+ * api/_lib/llm.js answered, e.g. "openai/gpt-oss-20b")* So the truth test is a MISMATCH between answered_by and model, not a prefix.* The first probe shipped prefix-only and read 0/225 fallbacks during a full* OpenRouter outage: a false negative from exactly the silence class this